Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 18 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.8428. Min Act: -0.0000

Data Index: 1236591 (C4 (Web Text))

Max Activating Token Index: 283

Text #1

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.5659. Min Act: -0.0000

Data Index: 973072 (C4 (Web Text))

Max Activating Token Index: 626

Text #2

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.6602. Min Act: -0.0000

Data Index: 594888 (C4 (Web Text))

Max Activating Token Index: 949

Text #3

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.2685. Min Act: -0.0000

Data Index: 741126 (C4 (Web Text))

Max Activating Token Index: 786

Text #4

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.5394. Min Act: -0.0000

Data Index: 725229 (C4 (Web Text))

Max Activating Token Index: 775

Text #5

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.9739. Min Act: -0.0000

Data Index: 59457 (C4 (Web Text))

Max Activating Token Index: 694

Text #6

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.2924. Min Act: -0.0000

Data Index: 428178 (C4 (Web Text))

Max Activating Token Index: 983

Text #7

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.1440. Min Act: -0.0000

Data Index: 1309509 (C4 (Web Text))

Max Activating Token Index: 252

Text #8

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.0928. Min Act: -0.0000

Data Index: 565782 (C4 (Web Text))

Max Activating Token Index: 646

Text #9

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.1134. Min Act: -0.0000

Data Index: 649895 (C4 (Web Text))

Max Activating Token Index: 699

Text #10

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.0047. Min Act: -0.0000

Data Index: 1264916 (C4 (Web Text))

Max Activating Token Index: 966

Text #11

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.8792. Min Act: -0.0000

Data Index: 1146575 (C4 (Web Text))

Max Activating Token Index: 639

Text #12

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.8666. Min Act: -0.0000

Data Index: 1319782 (C4 (Web Text))

Max Activating Token Index: 559

Text #13

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.8614. Min Act: -0.0000

Data Index: 208185 (C4 (Web Text))

Max Activating Token Index: 138

Text #14

Max Range: 4.8428. Min Range: -4.8428

Max Act: 4.0323. Min Act: -0.0000

Data Index: 937773 (C4 (Web Text))

Max Activating Token Index: 968

Text #15

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.8353. Min Act: -0.0000

Data Index: 871868 (C4 (Web Text))

Max Activating Token Index: 994

Text #16

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.8490. Min Act: -0.0000

Data Index: 96481 (C4 (Web Text))

Max Activating Token Index: 908

Text #17

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.8261. Min Act: -0.0000

Data Index: 901914 (C4 (Web Text))

Max Activating Token Index: 28

Text #18

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.8016. Min Act: -0.0000

Data Index: 1095194 (C4 (Web Text))

Max Activating Token Index: 740

Text #19

Max Range: 4.8428. Min Range: -4.8428

Max Act: 3.7701. Min Act: -0.0000

Data Index: 387390 (C4 (Web Text))

Max Activating Token Index: 937